Machine learning paradigm for structural health monitoring

Bao, Yuequan; Li, Hui

doi:10.1177/1475921720972416

Cited by 178 publications

(94 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ML originated from artificial intelligence (AI) and has been used in recent years in various scientific areas. Increasing applications of RS [48][49][50][51][52][53][54] and ML [55][56][57][58][59] in structural health monitoring of infrastructures in recent years motivates the authors to conduct this idea. Second, the authors intend to extend the proposed methodology for other types of bridges, including steel and stone (old) bridges.…”

Section: Indexmentioning

confidence: 99%

Application of Analytical Hierarchy Process for Structural Health Monitoring and Prioritizing Concrete Bridges in Iran

et al. 2021

View full text Add to dashboard Cite

This paper proposes a method for monitoring the structural health of concrete bridges in Iran. In this method, the bridge condition index (BCI) of bridges is determined by the analytical hierarchy process (AHP). BCI constitutes eight indices that are scored based on the experts’ views, including structural, hydrology and climate, safety, load impact, geotechnical and seismicity, strategic importance, facilities, and traffic and pavement. Experts’ views were analyzed by Expert Choice software, and the relative importance (weight) of all eight indices were determined using AHP. Moreover, the scores of indices for various conditions were extracted from experts’ standpoints. BCI defines as the sum of weighted scores of indices. Bridge inspectors can examine the bridge, determine the scores of indices, and compute BCI. Higher values of BCI indicate better conditions. Therefore, bridges with lower BCI take priority in maintenance activities. As the case studies, the authors selected five bridges in Iran. Successful implementation of the proposed method for these case studies verified that this method can be applied as an easy-to-use optimization tool in health monitoring and prioritizing programs.

show abstract

Section: Indexmentioning

confidence: 99%

Application of Analytical Hierarchy Process for Structural Health Monitoring and Prioritizing Concrete Bridges in Iran

et al. 2021

View full text Add to dashboard Cite

show abstract

“…With the development of advanced imaging devices and computer vision, vision-based structural monitoring and inspection have gained increasing attention in the field of civil infrastructure monitoring [ 10 , 11 , 12 , 13 , 14 , 15 , 16 ]. Computer vision techniques can be implemented into civil infrastructure applications to track the global motions/deformations of structures [ 17 , 18 , 19 , 20 , 21 ] and detect the local changes/damages [ 22 ] with the advantages of being non-contact, long-distance, low-cost, and less time-consuming and allowing automatic monitoring and inspection.…”

Section: Introductionmentioning

confidence: 99%

Pixel-Level Fatigue Crack Segmentation in Large-Scale Images of Steel Structures Using an Encoder–Decoder Network

Dong

Jin

et al. 2021

Sensors

View full text Add to dashboard Cite

Fatigue cracks are critical types of damage in steel structures due to repeated loads and distortion effects. Fatigue crack growth may lead to further structural failure and even induce collapse. Efficient and timely fatigue crack detection and segmentation can support condition assessment, asset maintenance, and management of existing structures and prevent the early permit post and improve life cycles. In current research and engineering practices, visual inspection is the most widely implemented approach for fatigue crack inspection. However, the inspection accuracy of this method highly relies on the subjective judgment of the inspectors. Furthermore, it needs large amounts of cost, time, and labor force. Non-destructive testing methods can provide accurate detection results, but the cost is very high. To overcome the limitations of current fatigue crack detection methods, this study presents a pixel-level fatigue crack segmentation framework for large-scale images with complicated backgrounds taken from steel structures by using an encoder-decoder network, which is modified from the U-net structure. To effectively train and test the images with large resolutions such as 4928 × 3264 pixels or larger, the large images were cropped into small images for training and testing. The final segmentation results of the original images are obtained by assembling the segment results in the small images. Additionally, image post-processing including opening and closing operations were implemented to reduce the noises in the segmentation maps. The proposed method achieved an acceptable accuracy of automatic fatigue crack segmentation in terms of average intersection over union (mIOU). A comparative study with an FCN model that implements ResNet34 as backbone indicates that the proposed method using U-net could give better fatigue crack segmentation performance with fewer training epochs and simpler model structure. Furthermore, this study also provides helpful considerations and recommendations for researchers and practitioners in civil infrastructure engineering to apply image-based fatigue crack detection.

show abstract

“…Zhou et al proposed model‐free hysteresis loop identification (HLA) algorithm based on the underlying mechanism, which is superior to model‐based identification 18,19 . Some approaches have been presented for the data‐based identification of model‐free hysteretic forces in structures, 20–23 especially, Bao et al 20 presented an excellent review on the state of the art of data science and engineering in structural health monitoring and their innovative approaches, Bao and Li 21 performed pattern recognition on quasi‐static monitoring data in high dimension feature space and proposed diagnostic approach through the variation of pattern parameters. The results were so promising that the study made the static monitoring data be useful and open a new field for structural health monitoring.…”

Section: Introductionmentioning

confidence: 99%

Identification of model‐free hysteretic forces of magnetorheological dampers embedded in buildings under unknown excitations using incomplete structural responses

Lei

Yang

Huang

et al. 2021

Struct Control Health Monit

View full text Add to dashboard Cite

Summary Magnetorheological (MR) damper is efficient to mitigate vibration of the structure subject to severe excitations. The identification of nonlinear characteristics of MR damper has attracted increasing attention. However, it is still challenging to identify model‐free hysteresis of MR dampers embedded in structures using only incomplete measurements of structural responses under unknown excitations. In this paper, an identification technique is proposed for this tough task, namely, nonparametric identification of MR nonlinear restoring forces in building structures under unknown excitations. The proposed technique involves identifications in two stages. In the first stage, the identification of linear parameters of bare structure and MR dampers under low‐level unknown excitations is conducted based on the generalized extended Kalman filter with unknown input (GEKF‐UI) by the authors. In the second stage, the hysteretic forces of MR dampers are proposed to be treated as “unknown fictitious forces” to the corresponding linear bare structure identified in the first stage. The generalized Kalman filter with unknown inputs (GKF‐UI) by the authors is adopted to identify all unknown inputs including the “unknown fictitious forces” originated from MR dampers. To demonstrate the proposed technique, some numerical examples are used to identify model‐free hysteresis of single or multi MR dampers with different nonlinear restoring force models in shear frames under unknown external force excitations or unknown seismic excitations, respectively. Furthermore, experimental testing of the identification of model‐free MR damper in a multi‐story shear frame under unknown external excitation is conducted.

show abstract

Machine learning paradigm for structural health monitoring

Cited by 178 publications

References 90 publications

Application of Analytical Hierarchy Process for Structural Health Monitoring and Prioritizing Concrete Bridges in Iran

Application of Analytical Hierarchy Process for Structural Health Monitoring and Prioritizing Concrete Bridges in Iran

Pixel-Level Fatigue Crack Segmentation in Large-Scale Images of Steel Structures Using an Encoder–Decoder Network

Identification of model‐free hysteretic forces of magnetorheological dampers embedded in buildings under unknown excitations using incomplete structural responses

Contact Info

Product

Resources

About